Short overview of 2.2 Intelligencer version enhancements.

New features

Last in Queue

“Last in Queue” behavior is basically an attempt to fight with very aggressive autoloaders (like symfony has) which aim to be the last in the queue by re-registering itself. “Last in Queue” behavior means that the manager will register itself once again if it’s not the last in the queue. You will have an ability:

* to specify number of attempts
* whether to do it only once or every time a new class is loaded

Two circles

“Two circles” behavior is devised to fight with unnecessary intelligencer cache invalidations and scanning. The first circle: go through all intelligencers; if the intelligencer is cacheable, load from cache. If the first attempt is unsuccessful the second circle takes place: go through all the cacheable intelligencers, invalidate cache and ask to load the class once again.

Cache for Templates

Templates, the same as Rules, will have an ability to cache the results. The good news is that templates will cache and invalidate particular classes not a list of them.

Persistent Cache improvements

Persistent cache will have an option to create all the subdirs that are missing.

This is a preliminary plan. The future updates can be found in Overview of Intelligencer 2.2

Intelligencer was released couple of days ago and hosted on google code.

Why should you use Intelligencer?

• You want to scan the contents of files and find all classes inside them. The framework provides the powerful instructions for searching with the cache support.
• You want to load the files according to the templates (e.g. Zend autoloading approach or Java class naming conventions)
• You want autoloaders to be prioritized
• You want your code to be independent from the locations of the sources

This framework is designed to prioritize and manage your class loadings.  There are two very powerful built-in itnelligencers: rules and templates. They are based on two different loading ideas. Basically, if you want the intelligencer to scan the files for the classes, use the rules. If you are able to create a path to the class without scanning the files (like Zend autoloading approach or Java naming conventions)

I will give you a short example of the code. The example below demonstrates how Rules Intelligencer works:

// Every time an attempt to load the class is made you want the
// intelligencer to check the cache and try to load from it.
// If an attempt is unsuccessful, the cache should be invalidated and #
// the files should be rescanned. 

    itManager::getInstance()
        ->setIntelligencerAndCreate(new itRulesIntelligencer())
            ->setCache(new itPersistentCache('cache.php'))
            ->setRulesAndCreate(new itComplexRules())
                ->setPrefix(dirname(__FILE__).'/drapeko/home/')
                ->plusDir('JustClass')
                ->getIntelligencer()
            ->control();

I hope you enjoyed it. Please visit official page http://code.google.com/p/intelligencer and find documentation and many useful examples.

One lite practicable idea for organizing version control tree for database development  is presented below:

The example of the svn tree is presented below:

/svn
  /features
    /user
	  /enable
	    create_user.sql
		create_user_permissions.sql
		pkg_user.pkb
		pkg_user.pks
	  /disable
	    drop_user.sql
	    drop_user_permissions.sql
	/comments
	 ...
  /scripts
	create_user_all.php
    drop_user_all.php
	recreate_database.php

The root directory consists of “features” and “scripts” dirs:

• Features – Logical functional units; a list of features we have (for example, a user or comments)
  • Each feature can be divided into “enable” and “disable” SQL scripts
    • Each enable/disable dir consists of the list of SQL scripts

• Scripts – List of scripts (php/perl/bash etc) that aggregate the features files into one sql file. The output of the script is the SQL file that contains all the SQL statements you need to enable/disable the feature.

DB developers work result is a list of SQL and pl/bash scripts for different variations (e.g. creates for empty db, patches et.

You have a straightforward list of directories. Your task is to create an hierarchical array of these directories. How can you do it? The simplest way is presented below.

You have an array of directories (straightforward list of directories):

	 $array = array(
		 '/home/drapeko/var',
		 '/home/drapeko/var/y',
		 '/home/drapeko',
		 '/home',
		 '/var/libexec'
	 );

And you would like to transform this array to hierarchy of directories:

 $array = array (
     'home' => array (
         'drapeko' => array (
             'var' => array (
                 'y' => array()
             )
         )
     ),
     'var' => array(
         'libexec' => array()
     )
 );

How can you do it?

First of all the below function will help us.

/**
 * This function converts real filesystem path to the string array representation.
 *
 * for example,
 * '/home/drapeko/var/y   will be converted to  ['home']['drapeko']['var']['y']
 *
 *
 * @param $path 		realpath of the directory
 * @return string		string array representation of the path
 */
function pathToArrayStr($path) {
     $res_path = str_replace(
          array(':/', ':\\', '/', '\\', DIRECTORY_SEPARATOR), '/', $path
     );
     // if the first or last symbol is '/' delete it (e.g. for linux)
     $res_path = preg_replace(array("/^\//", "/\/$/"), '', $res_path);
     // create string
     $res_path = '[\''.str_replace('/', '\'][\'', $res_path).'\']';

     return $res_path;
}

It simply converts the real path of the file to array string representation.

How can you use this function? I know it looks a little bit confusing. But it’s quite simple. Consider the example below:

 $result = array();
 $check = array();
 foreach($array as $val) {
 	$str = pathToArrayStr($val);
 	foreach($check as $ck) {
 		if (strpos($ck, $str) !== false) {
 			continue 2;
 		}
 	}
 	$check[] = $str;
 	eval('$result'.$str.' = array();');
 }
print_r($result);

Heh, how do you find it? This approach has helped me very much. I hope you will find it useful. :)

Several months ago I wrote an article that included several interesting examples of __autoload function, some autoloading approaches and a tiny script that is able find interfaces/classes and generate arrays of associations among these classes/interfaces, their locations and extended/included classes/interfaces.

http://wp.drapeko.com/2009/03/28/autoloading-in-php/

I widely use this script and I’ve found it really convenient.  I don’t have to think anymore of how to store my classes, what is the structure of the project etc. All this work is done by this script.

But… I have to launch it manually.

I’ve decided to go further and began to create a tiny framework called Intelligencer. This framework will extend functionality of the autoload generator script.

Some Intelligencers features:

1. It is very tiny and does not depend on other frameworks and external libraries. It’s very easy to integrate it.
2. It has an ability to store and control inheritance associations and relations between classes/interfaces and their locations.
3. If something has changed Intelligencer will automatically regenerate (if necessary) lists of associations. It’s really very useful in the development stage.
4. Intelligencer has a huge number of config settings. It’s flexible.
5. Intelligencer supports environments. You can easily create your custom environments and switch between them.
6. You can create several Intelligencers that will be responsible for different parts of your application.
7. You can work with Intelligencer both on config level and API level.

If you use it, it will do all work for you. It’s an open source. Location -  sourceforge.

The first release will come very soon.

I began to use Google Adsense two months ago. One month ago my account was banned. Unfortunately, the first appeal was unsuccessful and I did not get any explanations.  I don’t know the exact reason why I was banned. It makes me laugh but the number of ‘earned’ money was a little bit more than 10 dollars. But I’m really very interested in good relations with Google, that’s why several days ago I appealed for the second time.

I sent a letter below.

Dear Sir/Madam Appeal

I am writing for the second time regarding these misunderstanding. Your
previous letter made me very unhappy and I didn’t expect such a decision to
be made. I am indeed very interested in cooperation with Google and I am
convinced that there was a misunderstanding.

I claim that I have followed all the terms and conditions and I have not
broken any rules/ In particular, I have not place any forbidden content, I
didn’t click on the banners and I have not incited anybody to click on
them. What if somebody did it on purpose so as to deprive me? In that case
I have absolutely no protection against somebody’s bad intentions. My
account has been active only for a week. Is the statistics gathered during
such a short period of time sufficient to conclude this is indeed me who
tried to cheat Google?

I’d hate to lose the opportunity to work with Google and with AdSense in
particular. I have no intention to create other accounts under a different
name. I am not interested in illegal actions or unfair profit.  I am ready
to uphold my rights and prove that. I would really want to have my account
annulled so as to start building the long-term and trustworthy relations
with Google.

I appeal to Google to reconsider the matter.

I look forward to hearing from you,

Yours faithfully,
Roman

Have you been in a similar situations? What is the result?

P.S. I decided to be independent from one particular adv company. I will try to implement one interesting idea very soon in terms of advertisement. Stay in touch. :)

Situation: you would like to connect remotely using SSH protocol to Linux/Unix OS from Windows environment and to run remotely a program that uses X-Windows (e.g. Eclipse) .

How can you do it?

1. Download SSH client. The most popular is free PuTTY client (http://www.chiark.greenend.org.uk/~sgtatham/putty/)
2. Secondly download and install X-Windows server for your Windows environment. I suggest to use xming (http://sourceforge.net/projects/xming). xming installation tips are available here: http://gears.aset.psu.edu/hpc/guides/xming/
3. Open Putty, go to Connection->SSH->X11 and enable X11 forwarding



4. Connect to the remote host and launch the program

I took Oracle beta 1z1-146 exam in the end of January, 2009. Several days ago I checked prometric.com and discovered that this exam was passed. Now I’m one of the first Oracle Advanced PL/SQL Developer Certified Professional all over the world!!! it’s my second OCP :)

I did not find any dumps, any mock exams or any sample questions, while I was preparing for this exam. I used only these http://rapidshare.com/files/127182353/D52601GC10_netbks.com.rar official Oracle preparation slides (remember to switch comments – it’s a book). 1z1-146 exam was really tough. 165 questions in 180 minutes. Production exam requirements: time – 90 min, questions – 68. There is a difference, is not it?

If you are going to take one of Oracle beta exams and you have time and ability to stay calm, you are not interested to economize some money I suggest waiting for a production one. Pluses: 1) You will probably get results two weeks earlier then beta-takers 2) More time, less questions 3) It’s not so  stressful 4) The probability to pass is much higher.

Do you agree? What is your experience in taking certification exams? :)

How to install Rational Rose on Windows XP Home Edition?

If you are trying to do it, you probably get the following message:

“We are attempting to install on an unsupported operating system. We recommend that you install on a supported operating system. See your Rational product’s Release Notes for a complete list of supported operating systems and service packs.”

Fortunately, this restriction is artificial one. You can turn it off.

1. Start -> Run -> cmd
2. Go to the directory where Rational is installed (example, cd C:/Program Files/Rational)
3. Find Rose.msi file. Usually it’s located in the SETUP folder. It can be called 1041_Rose.msi or something very similar.
4. Type msiexec.exe /l*vx inst.log /i Rose.msi /c DISABLE_PLATFORM_BLOCKS=1

A month ago I relocated to London. And now I’m looking for a job here. The process is quite boring. Agents, sites, applications, cover letters, CVs. And.. unfortunately no visible result yet. But sometimes potential employers ask to do interesting tasks. For example, one of them asked to answer the questions below:

My version could be found below.

What is your favorite

1. programming language

My experience includes programming in Java, JavaScript, C, C++, Pascal and database inner languages (PL/SQL, SQL PL etc). Still, my favorite language is PHP. It’s small and fast. It’s simple and convenient. Have you ever used arrays in PHP? They are divine.

2. Operating system

I like to use the most modern software. My computer likes to be up-to-date. My devices like to be identified from the first attempt. Friendliness and usability are the most important things for me. I partook in action called “A year without Windows, work on Linux”. It was a prolonged battle but I was responsible and bet on, so I installed Windows on 366 day, not earlier. I choose Windows, but I really like bash in Linux/Unix.

3. Editor/IDE

• Notepad++ is for rapid notes and corrections. Light and powerful.
• Netbeans is for Java. I have not seen a better one for Java.
• Aqua Data studio is the best one for database development. Works with all possible databases: from H2 to Oracle.
• Eclipse is for all other projects. Millions of plugins. Thousands of good plugins.

4. Version control system

I have recently switched from CVS to SVN and I’m not thinking about going back. SVN manipulates all file types, works faster and is still alive.

5. Bug tracking system

I have used ClearQuest and Jira. I don’t have any specific preferences, but the first one exists in several realisations and could be easily integrated into Eclipse environment.

6. Development tool

IBM Rational package is highly professional software. Programs extremely well integrate into each other. You could design even an aircraft within this environment. Mostly I like Rational Rose Data Modeler and its’ forward and reverse engineering abilities.

7. Web site

Favorite websites are constantly and dynamically changing. When I’m heavily working within Oracle and PL/SQL environment, the most favorite site is http://ora-code.com/. When I’m working with PHP – http://php.net and Java – http://java.sun.com/javase/6/docs/api/.

I visit regularly bbc.com and economist.com websites to catch last world news.

8. Movie

The Green Mile. Emotional. Strong ideas. Great actors.

9. Record

I graduated musical school on the saxophone. I love jazz. Check that out http://www.youtube.com/watch?v=XPxSSBe8DaU. Is not it wonderful?

10. Book

Allen Carr – The Easy Way to Stop Smoking. If you have faced a smoking problem, I highly recommend you this book.

What do you *think* of

1. XML

XML is very simple and powerful. When I was writing my bachelor’s thesis called XML Data Search in Relational Databases I run into one of the Microsoft executive’s opinion. He thought that XML’s popularity was fabricated and would be short-time frame. He told it in 2002. Was he mistaken? Now XML is everywhere. XML is in your telephone, is a foundation of your favorite feed, in configuration files and it is even a basic format for Microsoft Office documents. Unbelievable, but XQuery is a primary language in DB2 database. Oracle also keeps this realization in plans. SQL/XML is 14th extension of SQL standard.

But history does not remember occasions when the thing becomes universal and is used everywhere. XML has one big minus. The ratio of all information to useful information is approximately 10:1 in the average XML file. Have you heard abou Json, Yaml? I have. And I think the computing world will probably hear about them in the nearest future. What if that executive was right?

2. Web Services

What would I choose: remote procedures, messages or resources? I like the last two. I strongly believe that services should “understand” their responsibilities very deeply and “do” their whole job very well. I like to think in abstractions. For me it’s not important at all what realization on the server side is. I need to be convinced that provision of this service is somebody else’s responsibility. I don’t want to depend on remote libraries.

3. Flash

As far as traditional plain web application is concerned, XHTML is for a structured data, CSS is for presentation, JavaScript is for actions and effects, Flash is for

• Applications such as games, webcasts, movies or records. As you know JavaScript cannot work with client files. Sometimes flash applications could be used very efficiently, for example, when you want to create a web application that would resize user pictures. If you use JavaScript you will realize this service on server side, but if you use Flash you can do it on the user side. What is the plus? Traffic.
• Advertisements. I presume that flash-movies will be very intensively used as advertisements in the nearest future.

But if we speak about the future technologies, I think there is a very huge probability that we will see quite many different visualization technologies in the future. Not only flash. For example, Microsoft stake on its own technology called Silverlight.

4. Excel

Special language was designed for excel templates in my previous job. Fragments of the code were integrated into excel comments. An application that dealt with these templates was located on the server side. What was the plus? We had the ability to develop very complex excel reports without using any special environment but excel.

You could think about excel in different ways. J Excel could be really powerful not only in traditional way

5. AJAX, “web 2.0″

Two years ago one could have claimed that Ajax was the future of the web. Today it appears to be a misbelieve. Ajax is the present, but probably not the future. Why? It’s very efficient.

I would like to say that I constantly use Google Docs. The idea to be independent from the computer device is a good one. I like it very much. Now we can manage and modify our files through the internet, tomorrow we will use Web operating systems to install the programs. Ajax will help us very much.

6. the Semantic Web

What is the purpose of the Semantic Web? To have the ability of describing everything. To have the ability to query everything. I think that the semantic web is in a same position that Ajax was in two years ago. Associations between friends in social networks, OpenID, RSS, Yahoo Answers, Google search opportunities are just the first steps in developing one large semantic web.

7. Agile Development

I like it.

8. UML

UML is a really good layer between the ideas and realization, between the analysts and developers.

I like forward-engineering. Moreover I think that in the nearest future most of the code will be generated by moving different components or objects on the screen. Java will become a low-level language. C will become an underground level language. UML and forward-engineering are just one of the first steps.

9. OO

I like to think in abstractions. I programmed in Java, PHP OO, C++ and used Oracle OO capabilities.

UML is just a modeling language, it’s just a tool that helps realize several conceptions. UML helps visually realize the conception of OO paradigm. OO is an excellent layer between the real life and computer.

10. Microsoft, Google and Apple

I think Microsoft will try to overtake Google in Web.
Apple and Google will try to overtake Microsoft in browsers.
Apple will try to overtake Microsoft in Operating system.
Google will release the first web operating system.

What is your version?

What do you think? You are welcome!